Draft for discussion 5 July 2004
THE RATIONALE FOR MULTILATERAL LENDING A CRITICAL ASSESSMENT Yilmaz Akyüz1
I. Capital market failures and multilateral lending
Multilateral lending has been a central element in international economic cooperation since the establishment of the Bretton Woods Institutions (BWIs) towards the end of the Second World War. These institutions in general and their lending activities in particular were designed to address specific challenges confronting the architects of the postwar international economic order. These included the need, on the one hand, to avoid the repetition of the difficulties that traumatised the world economy during interwar years, and on the other hand, to repair the damages inflicted by the war on the productive capacity and living standards of the countries involved. Accordingly, two distinct objectives were attached to multilateral lending: restoration and maintenance of international economic stability, and reconstruction and development.
The world economic and political landscape has undergone significant changes since the Bretton Woods Conference (BWC), including the end of colonialism, the rise
1 Second holder of the Tun Ismail Ali Chair in Monetary and Financial Economics, University of Malaya, Kuala Lumpur; and former Director of Division on Globalization and Development Strategies, and Chief Economist, UNCTAD. Paper prepared for the UNDP Project “Public Finance in a Globalizing World”. I am grateful to Richard Kozul-Wright, Jan Kregel, Inge Kaul, Ron Mendoza and two anonymous reviewers for comments and suggestions. None of the persons mentioned are, of course, responsible for any errors or omissions.
2 and the fall of communism, closer global economic integration, and a rapid development of international financial markets. The BWIs have responded to these changes by realigning their activities both in multilateral lending and policy surveillance, though often in an ad hoc manner. Consequently, the objectives pursued by these institutions and the modalities of their operation are now significantly different from their original design.
Simultaneously, there has been an intense debate over the role of the BWIs and the need for multilateral lending. It has been increasingly suggested that multilateral lending has lost much of its rationale as a result of the rapid development of international capital markets and increased access of developing countries to external private financing (see, e.g. Walters 1994; Meltzer Commission 2000). This view has found support particularly among those who subscribe to the neo-liberal view and see multilateral lending as unnecessary and even harmful intervention with the efficient functioning of international capital markets, emphasizing that governments fail more often than markets in supporting stability and growth. Ironically, these institutions, notably the World Bank, now appear to be disarmed, as lenders, by their own logic, since this neo-liberal view constitutes the intellectual basis of the so-called Washington Consensus promoted by the very same institutions.
Multilateral lending is a form of global collective action designed to provide external financing to countries in pursuit of certain agreed objectives. The economic literature has established that collective action, rather than private action, would be desirable under two sets of circumstances. First, when markets fail to generate efficient outcomes due to reasons such as the presence of externalities, public goods, imperfect and asymmetric information, monopolies or imperfect competition, and incomplete markets. This is the Pareto efficiency argument wherein there is scope to improve the
3 conditions of some without worsening those of the others. Second, when there is a broad agreement on certain social and humanitarian objectives that cannot be guaranteed even under efficient outcomes. Collective action may be required for both efficiency and humanitarian reasons since “there are certain market failures which not only lead to inefficiencies (Pareto inefficient outcomes) but the incidence of those inefficiencies bears disproportionately on the poor” (Stiglitz 2003, p. 2).
The main reason why the international community should be concerned with the provision of external finance to countries rather than leaving it to international capital markets is that these markets fail to fulfil this task adequately and this failure has ramifications not only for the countries concerned but also for the international community as a whole because of existence of international externalities and global public goods.2
International externalities operate through various channels. When a country facing a tightened payments constraint as a result of trade shocks or increases in international interest rates is forced to cut imports, its trading partners would also suffer because of reductions in their exports and economic activity. Again liquidity crises in debtor countries could create problems for their international creditors, threatening to destabilize international financial markets. Similarly, environmental, health and security problems associated with widespread poverty in poorer countries tend to generate negative externalities for the international community as a whole, necessitating provision of development finance to address the root causes of these problems in the countries concerned.
2 For an earlier discussion of “externalities and international public goods ... in the economic interactions of nations” see Bryant (1980, pp. 473).
4 The increased significance of international externalities associated with growing interdependence among countries, and in particular the problems posed by repeated bouts of financial crisis in developing countries, have been important factors in the increased attention paid to global public goods in the recent debate on global governance and collective action (Kaul, Grunberg and Stern 1999; and Phillips and Higgott 1999). What exactly constitutes global public goods seems to be highly contentious. However, the spectrum has been widening given the tendency of economics to invade and colonize other areas of social science, extending the notion of public goods into spheres traditionally dealt with by political scientists and sociologists (Kindleberger 1986, p. 1). Indeed, global security, global economic stability, global environment, knowledge, humanitarian assistance and global health have been included among global public goods in recent discussions (Kaul, Grunberg and Stern 1999; Stiglitz 2002a).
The pursuit of global humanitarian objectives and the provision of global public goods often require provision of finance to poor countries in the form of grants or highly concessional loans and technical assistance rather than multilateral lending of the kind that could be provided by capital markets. In this case global benefits attached to redistribution from richer to poorer countries rather than international capital market failures would constitute the rationale for multilateral financing (Gilbert, Powell and Vines 1999, p. 607). By contrast, the rationale for non-concessional multilateral lending should be sought primarily in international capital market failures, even though distributional considerations also come into play in such lending since it often contains a certain element of subsidy.
A major reason why rational behaviour in unfettered private capital markets give rise to inefficient outcomes is the presence of imperfect and asymmetric information (the borrower knows more about his investment than the lender, and information costs about
5 borrowers’ riskiness are high) and incomplete contracts (lenders cannot control all aspects of the borrower’s behaviour). These problems imply that lenders are generally unable to assess the quality of a loan and defaults are costly. On the other hand, as loan rates are raised to cover risks, the average quality of loans will fall because of adverse selection (lending to high-risk borrowers willing to pay high interest rates) and moral hazard (inducing “good” borrowers to invest in riskier projects). As a result, the expected rate of return net of default will decline once the loan rate has reached a certain level. Beyond that level, the lender would be inclined to ration the borrowers rather than push up the loans rates until the market is cleared. In other words, credit rationing would be an equilibrium outcome of rational behaviour of lenders under conditions of imperfect and asymmetric information and incomplete contracts.3
Clearly, the problems of information and incomplete contracts are more acute for international lending to both public and private sectors in developing countries which lack sophisticated financial institutions and markets, and effective disclosure requirements, accounting standards and prudential regulations. These lead to the overestimation of risk, increases in risk premia and loan rates, shortening of maturities, and severe credit rationing. Multilateral lending could thus rectify such capital market failures by extending long-term loans at below-market rates, including to countries which have little or no access to international capital markets. Moreover, to the extent that multilateral financial institutions (MFIs) have access to better information than private lenders and are able to exert more effective control over the behaviour of borrowers, such lending would not necessarily be associated with higher risk of default.
The second manifestation of capital market failure is instability in the volume and
3 This analysis of credit rationing was first developed in a seminal paper by Stiglitz and Weiss (1981). For a detailed discussion see Davis (1992, chap. 1) and Mishkin (2004, Part III).
6 terms of international private capital flows, including boom-bust cycles and currency and debt crises that have occurred with increased frequency in emerging markets in recent years. These reflect abrupt and unexpected changes in markets’ assessment of risks which cannot always be attributed to policy shifts. A plausible explanation is provided by the theory of endogenous financial fragility developed by Minsky (1977), which sees financial cycles as an intrinsic feature of market economies.4 Booms generated by improved opportunities for profitable investment lead to underestimation of risks, overexpansion of credits and over-indebtedness. Excessive risk taking eventually results in a deterioration of balance sheets and increases in non-performing loans. Lenders respond by reassessing risks and sharply cutting credits, which in turn lead to credit crunch, debt deflation and defaults. However, while crises are almost always associated with a certain degree of financial fragility, they can also take place in the absence of serious economic weaknesses, because of the so-called self-fulfilling prophecies resulting from the existence of multiple equilibria and debt runs associated with herding behaviour and collective action problems (Obstfeld 1996; Krugman 1996, 1998, 1999; Eichengreen and Portes, 1995, pp. 8–9).
All these mean that financial markets tend to behave pro-cyclically and countries may not have access to international liquidity when they need it most. Accordingly, multilateral lending can be needed to counter fluctuations and boom-bust cycles in international private capital flows. Such lending could potentially serve two distinct purposes: to allow countries to finance their imports of goods and services and factor income payments (current account financing); or to help them maintain capital account convertibility and to stay current on their external debt repayments (capital account financing).
4 The approach goes back to Fisher’s analysis of the Great Depression: see Davis (1992, chap. 5) which also gives a survey of various explanations offered for financial instability and crises.
7 However, it is also important to keep in mind that capital markets failures do not always necessitate multilateral lending. It may in principle be possible to address such failures through other forms of collective action rather than direct multilateral lending. For instance, an important part of the recent debate on the reform of the international financial architecture has focussed on whether private capital flows could be made more stable and hence reliable sources of international liquidity and development finance (Akyüz 2002a). Clearly, to the extent that this could be achieved, the rationale for multilateral lending would be weaker. Similarly, to the extent that MFIs help reduce information imperfections and asymmetries, improve risk assessment, share in risks or provide guarantees on private loans, their reliance on multilateral lending would be lessened in rectifying capital market failures.
It should also be noted that official financing can be arranged bilaterally rather than multilaterally. However, there is a wide agreement over the superiority of multilateral lending since bilateral lending is often driven by political considerations. (Cassen 1994; Kapur and Webb 1994; Rodrik 1995; Gilbert, Powell and Vines 1999). In fact the introduction of donor-driven concessional windows in the BWIs is said to have altered “the shield against political interference” (Kapur and Webb 1994, p. 232). This issue will not be taken up in this paper, and it will be maintained that to the extent that there is a need for official financing, it should be organized on a multilateral basis.
The following section will examine the original rationale for multilateral lending and the mandate and design of the BWIs. This is followed by an account of the evolution of multilateral lending and the role and objectives of these institutions since the War. Section four will discuss the extent to which private lending to developing countries has manifested the kind of behaviour examined above, and assess the extent to which international financial markets can be expected to meet the financing needs of
8 developing countries today. The paper will conclude with a discussion of the rationale and objectives of multilateral financing today, including an assessment of various proposals for reform.
Throughout the discussions a distinction will be made among three types of multilateral lending. First, counter-cyclical provision of liquidity for meeting temporary current account imbalances. Second, development finance including both grants and non-concessional multilateral lending. Third, lending to countries facing capital-account crises in order to bail out private creditors. The paper concludes that the rationale for counter-cyclical lending is now stronger in view of the increased instability of the international economic environment, greater vulnerability of developing countries to external shocks and pro-cyclical behaviour of international financial markets. In development finance, the rationale for multilateral lending is weaker today than in the immediate postwar era, but still continues to be valid for a large number of developing countries. By contrast the rationale for development grants is now stronger than was originally envisaged. Finally, the paper finds no strong rationale for crisis lending to bail out private creditors: the original considerations that precluded IMF lending to finance capital outflows continues to be valid today since bailout operations do not correct but aggravate market failures by creating moral hazard, and shifting the burden of financial crises disproportionately on to the poor.
II. The original design
The main objective pursued by the architects of the postwar international economic system was to avoid the recurrence of a number of problems that had led to the breakdown of international trade and payments in the interwar period and contributed to
9 one of the bloodiest armed conflicts in the history of mankind.5 These problems were due primarily to the failure to make provisions for postwar reconstruction and potential balance-of-payments difficulties. Excessively onerous debt payments were imposed in the Treaty of Versailles, and there was also an insistence on inter-allied debt repayment. Efforts by each country to put its house in order through unilateral and mutually inconsistent stabilization plans under conditions of excessively volatile short-term capital flows encouraged by disparities between national inflation and interest rates gave rise to misaligned exchange rates and generated frictions in the trading system. Increased financial austerity and instability eventually culminated in the Great Depression, creating pressures for governments to protect employment and to support domestic price levels through beggar-my-neighbour exchange rate policies, and trade and payments restrictions introduced on a quid pro quo basis.
The plans to address potential postwar problems had already begun in the late 1930s when the war was still raging. As “peace was seen as linked with world prosperity, and prosperity, with free trade, free capital movements, and stable exchange rates” (Mikesell 1994, p. 4), efforts were concentrated on establishing conditions for global economic stability and security, and for sustained and broad-based growth in incomes and employment. These required not only dismantling the ad hoc exchange controls and discriminatory trade barriers introduced in the interwar period, but also establishing additional safeguards to ensure efficient operation of the international economic system. Mindful that the interwar economic disintegration was due to the fact that the recovery from the First World War was based on shaky ground which contained seeds of economic imbalances and political conflict, the postwar planners strived to establish a process of reconstruction and global economic reintegration based on three
5 For a brief account of these problems see Oliver (1975, chap. I) and Dam (1982, chap. III).
10 lessons drawn from the interwar years:
$ Market failure: economic problems could not all be left to markets which can fail to bring about economically efficient and politically acceptable solutions;
$ Interdependence: solution to problems in areas such as trade, finance, investment, reconstruction and development should not be sought in isolation because of close linkages among them;
$ International cooperation: nor could solution to such problems be left to uncoordinated individual country actions because they could lead to inconsistencies and conflicts. Furthermore, no country should be expected to put its house in order irrespective of global economic conditions.
Thus, the restoration of a stable global economic system required a shift from purely national policy formulation to a multilateral system based on the recognition of economic interdependence, need for cooperation and supportive multilateral institutions. The process of creating a coherent set of international institutions in support of global stability and prosperity could not be completed,6 but agreement was reached on creating two distinct institutions to deal with postwar financial problems: the IMF and the IBRD. Although the arrangements agreed at the BWC contained a long-term framework for an international financial system, the main concern was to address the potential problems expected to emerge in the immediate postwar era.7
6 Two institutions that were to complement the BWIs in support of expansion of trade and prosperity under conditions of stability, The International Trade Organisation and The International Commodity Stabilization Fund, were not created.
7 This was certainly the case with White’s proposal (Mikesell 1994, p. 7). On the other hand, Keynes was more
11 A. The original rationale of IMF financing
The interwar experience suggested that prevention of disruption of the international monetary system and restoration of open international trade were essential for an orderly transition to durable peace. One of the factors that contributed to the breakdown of the international monetary and trading system in the interwar period was lack of international cooperation and multilateral discipline over exchange rate policies. As noted, uncoordinated currency stabilization policies in the wake of the war gave rise to misaligned exchange rates on the basis of which countries returned to the gold standard. With inappropriately fixed exchange rates, sustainable trade balances could not be attained under conditions of rapid growth. This problem was aggravated by the highly unequal international distribution of gold reserves. Countries facing persistent deficits were thus forced to undertake deflationary adjustment or resort to exchange restrictions on current transactions because of the absence of any obligation on the side of the surplus countries to share the burden of adjustment, as well as any multilateral arrangement for the provision of international liquidity. Deflationary adjustment and exchange restrictions in turn generated negative externalities and led to frictions in international trade. When eventually the gold standard was abandoned, there was widespread resort to competitive devaluations and ad hoc trade and exchange restrictions, which created even greater disruptions to the trading system.
For postwar planners the obvious lesson was that exchange rate stability and sustained expansion of output and employment were essential for avoiding tensions and disruptions in international trade. This, in turn, required global arrangements based on three ingredients: first, multilateral discipline over exchange rate policies; second, mechanisms for the provision of international liquidity; and third, restrictions over interested in the immediate balance of payments problems than with long-term investment (Oliver 1975, p. 154).
12 destabilizing capital flows. Thus, the IMF was created to ensure an orderly system of international payments at stable but multilaterally negotiated, adjustable exchange rates under conditions of strictly limited international capital flows. Its most important function was to provide international liquidity not only to avoid deflationary adjustments and trade and exchange restrictions in deficit countries, but also to help maintain stable exchange rates in the face of temporary payments disturbances.
With the failure to establish an international commodity stabilization fund to promote commodity price stability, the responsibility for dealing with fluctuations in export earnings of developing countries effectively fell under the IMF’s role for the provision of short-term liquidity. However, the Fund was designed primarily for securing the stability of external payments and exchange rates of the major industrial countries, rather than for the stabilization of balance of payments of developing countries. Indeed, the United Kingdom was expected to face serious payments difficulties after the war, while many developing countries, notably in Latin America, were enjoying a boom in export prices and running trade surpluses due to a strong demand for primary commodities.
The modalities of provision of liquidity were perhaps one of the most controversial issues in the negotiations leading to the BWC, particularly between the two key architects of the postwar system, White and Keynes. The plans independently prepared by White and Keynes both provided for international liquidity to enable countries to stabilize their currencies, but Keynes’s plan for an International Clearing Union went much further, effectively proposing that the reserves of surplus countries should be automatically available to deficit countries for meeting their current account needs (Mikesell 1994, pp. 12-13; Dam 1982, pp. 77-81; and Oliver 1975, chap. VI). At the end White’s scheme prevailed and an agreement was reached on establishing a fund
13 with contributions from countries partly in gold and partly in their own currencies, which would be available for drawing by those in need of international reserves. The Fund was also empowered to borrow from member countries to supplement its reserves in currencies in large demand.
It is notable that “the language of credit (repayment, interest and the like) is missing from the Articles of Agreement. ... One of the basic concepts of the Fund is that balance-of-payments assistance takes the legal form of a purchase of a strong currency for the member’s own currency rather than the borrowing of the strong currency. Similarly, when the time to reverse the transaction occurs, the member repurchases its own currency” (Dam 1982, p. 106). This formulation had the connotation that a member would have the right to determine how much it would draw, treating its subscription as its own reserves. Indeed, Keynes strongly argued that members should have unconditional access to the Fund within the limits of their quotas, and that “it would be very unwise to try to make an untried institution too grandmotherly.”8 However, unconditional drawings were generally resisted by the United States on grounds that it would be the only possible source of net credit in the immediate postwar era with the dollar being the only convertible currency at the time. Eventually an agreement was reached on a compromise language in Article V which stated that “a member shall be entitled to buy the currency of another member from the Fund in exchange for its own currency”, and “required” the purchasing member to make a “representation” that the other member’s currency was needed “for making in that currency payments which are consistent with the provisions” of the Articles. Most countries believed that this formulation gave the right to members for unconditional drawing within the limits of their quotas, and gave no right to the Fund to challenge a member’s representation (Dell 1981, pp. 4-5), but there was considerable
8 See Dam (1982, pp. 106-108) and Dell (1981) for a discussion of the issues involved, and the Anglo-American debate on conditionality.
14 room for interpretation.
During the negotiations for the Bank there was considerable debate on whether the task could be effectively performed by private lenders, but this was not the case for IMF financing. Although there were some instances of currency stabilization during the interwar period supported by officially arranged private lending (Oliver 1975, pp.12-15), in the debate on the postwar economic architecture it was almost taken for granted that commercial banks could not be relied on for such a task, particularly given the high degree of volatility displayed by short-term international capital flows during interwar years.
Access to Fund liquidity was designed for current account transactions alone. Article IV stated that one of the essential purposes of the international monetary system was to provide a framework facilitating the exchange of capital, a statement which is included among general obligations regarding exchange arrangements. However, the more specific references to capital movements in Article IV permitted recourse to capital controls so long as they did not restrict payments for current transactions, and actually gave the Fund the authority to request a member country to impose controls to prevent the use of funds from its General Resources Account (GRA) to finance large or sustained capital outflows. In doing so, the arrangements were in effect discouraging reliance on private financial flows for balance of payment financing.
The Fund’s sources of assets, members drawing rights to those assets as well as the voting rights were all linked to a single concept of quota (Dam 1982, pp. 101-102). The quotas were determined through a highly politicized exercise so as to give a veto
15 power to the United States over key decisions.9 From an economic point of view, using the same criteria both for countries’ contribution to the Fund and for their drawing rights might have assured sufficient funding in reserve currencies through contributions by major industrial countries, but it also implied that drawings by poorer countries with smaller quotas would be highly restricted compared to their needs. As correctly argued, the “use of a single Fund quota to serve three purposes was both illogical and unnecessary... There could well have been one quota based on, say, foreign trade and export variability to govern drawing rights, a second quota based on reserves and balance-of-payments history to govern contributions to the Fund, and a third quota based on economic and political importance to determine voting rights” (Mikesell 1994, pp. 37- 38).
B. The original rationale of IBRD lending
The IBRD was established primarily with a view to assisting in the postwar reconstruction “by facilitating the investment of capital for productive purposes”. Although assisting in the “development of territories of members”, including “the encouragement of the development of productive capacities and resources in less developed countries” is mentioned among its purposes in Article I, financing development was a secondary objective, almost an afterthought. Indeed, at the time, few developing countries existed with independent status. Most Europeans saw a trade-off between reconstruction and development financing, and emphasized the urgency of
9 According to Raymond Mikesell, who was actually assigned the task of calculating the quotas, “(a)ssigning quotas in the Fund was the most difficult and divisive task of the conference... The quota formula was not distributed, and White asked me not to reveal it.... I tried to make the process appear as scientific as possible, but the delegates were intelligent enough to know that the process was more political than scientific”. Mikesell (1994, pp. 35-36); see also Woods (1998).
16 projects in war-torn areas. Attempts by Latin American countries to secure a balance between the two met with resistance, and Article III settled with a compromise that there “should be equitable consideration to projects for development and projects for reconstruction alike” (Oliver 1975, pp. 189-90).
While postwar reconstruction of productive capacity required substantial imports, Europe had neither financial resources nor export capacity to pay for them. The war-torn countries thus faced a structural trade imbalance that required long-term resource transfers far larger than the liquidity envisaged for the IMF. Although this was also the case for development finance, there was an important difference: postwar reconstruction was primarily a transitory problem, which could be completed, given the necessary financing, in a relatively short period of time, since the required complementary skills, know-how, infrastructure and institutions were largely in place. This was not the case in much of the developing world. The postwar planners paid little attention to this difference, in large part because of their focus on reconstruction, but in the event the IBRD would become more prominent as a development agency.
In carrying out its objectives there was a consensus that the IBRD would not compete with private investors but, as stated in its Article I, provide finance only “when private capital is not available on reasonable terms.” While Article I emphasized guaranteeing private investment rather than direct lending, it was widely believed that private capital would not be available in the volume required for a rapid reconstruction as well as for meeting the needs of developing countries. However, the rationale for IBRD lending was not simply the inadequacy of the amount of private finance that would be forthcoming or lack of depth in international capital markets. This is clearly indicated by the reference to the non-availability of private capital “on reasonable terms”. Indeed, a major source of concern was that the terms and conditions of private financing, notably
17 market interest rates, would not be appropriate for the conditions prevailing in the borrowing countries. The United States was particularly emphatic regarding capital market failures in meeting the objectives sought by postwar planners:
“White was convinced that private investors could not be relied upon to provide the capital that would be needed for postwar reconstruction. He also felt that even after the postwar transition period, the normal flow of capital from rich to poor could not be left solely to the private investment markets of the world. The lessons of the twenties had been that long-term private capital movements tended to enforce, rather than mitigate, the spread of international business fluctuations and that the high interest rates and the relatively short-term maturities of private portfolio investments tended to make unproductive what might otherwise be productive international ventures” (Oliver 1971, p. 4).
The financing to be provided by the IBRD was to be in loans rather than grants. However, in two respects these loans differed from private lending. First, the Bank loans were to carry lower rates than charged by private lenders. This would be possible not only because its paid-in capital provided interest-free loanable funds, but also the Bank would be charging only a small commission over its borrowed funds in lending to its members. Thus, compared to private lending, there was an element of a subsidy in Bank loans. Second, unlike private markets which charged different rates for different borrowers on the basis of differences in perceived risks, the Bank would charge the same rate of interest and a flat commission to all borrowers.
In the discussions leading to the BWC, it was often argued that Bank lending for a project in an individual country could create positive externalities for the world economy as a whole. Thus, White “hoped that, by reducing interest rates, by providing for relatively longer repayment periods, and perhaps by helping to diminish worldwide business fluctuations and short-term variations in relative prices, some other unprofitable projects might be profitable” (Oliver 1971, p. 40). Keynes took this even further, suggesting that the Bank should not “seek to avoid the incurring of some measure of
18 reasonable risk ... where the loan is in the general interest of reconstruction or developing the world’s resources or expanding international trade along mutually advantageous lines; ... [it should] seek to conduct its operations taken as a whole... rather than seek full security from risk in each transaction taken separately” (Oliver 1975, p. 193). Even though such provisions were not included in the Articles, the Bank was clearly expected to finance projects that would not be found profitable by financial markets but beneficial to the world as a whole.
While Article III explicitly stated that loans made or guaranteed by the Bank shall be for the purpose of specific projects of reconstruction or development, exceptions could be granted in special circumstances. This left discretion to its Board for loans for currency stabilization and general balance of payments. Although there were no explicit provisions for counter-cyclical use of the Bank’s loans, this was not altogether ruled out as it was agreed that the Bank would conduct its operations with due regard to effects of projects on business conditions. The Article also prohibited tied-aid: that the Bank should impose no conditions that the proceeds of a loan must be spent in the territories of any particular member or members.
A dilemma was faced in reconciling the means with the objectives of the Bank; that is, its capital base would need to be provided by the very same countries it was designed to help in their reconstruction and development. This was resolved by an agreement that provided that each member country would pay only 20 per cent of its subscription to the Bank’s capital, with the rest being callable as the Bank ran out of resources (paid in capital plus reserves) to meet its obligations on funds borrowed from international markets.10 This guarantee provided by its shareholders greatly helped the
10 When the subscriptions were doubled in 1959 the required paid-in capital was reduced to 10 per cent, allowing the Bank even greater leverage in borrowing from bond markets.
19 Bank in subsequent decades in raising funds at highly favourable terms, thereby introducing an additional subsidy element into its loans and reducing the cost to its borrowers.
III. The postwar evolution of multilateral lending
A. IMF lending
The past sixty years have seen a constant distancing of the IMF from the objectives set in the BWC, particularly with respect to automaticity of counter-cyclical current account financing. This is perhaps best seen in the evolution of the conditions governing the members’ drawings from the Fund and the operational procedures followed.
The process of legitimizing and ratcheting conditionality started soon after the BWC.11 In this respect, the May 1947 decision of the IMF Board on the meaning of the word “represents” in Article V constitutes a turning point. In this decision the Board explicitly ruled against any concept of members’ right to drawing from the Fund, and gave a green light to the IMF to challenge “representations” by members, thereby discarding the notion of automaticity in drawings on the Fund. This was followed by a key decision in 1952 which formally adopted conditionality, and stipulated that the Fund’s position vis- à-vis a request by a member would depend on whether the problem is of a temporary nature and whether the policies proposed would be adequate to overcome the problem within such a period. In the same decision standby arrangements were introduced as the central operational modality (IMF 1969, Vol. 3. pp. 228-230).
11 For an excellent account of the rationale and evolution of IMF conditionality see Dell (1981). See also Jungito (1994), and Buira (2003).
20 A few years later in 1956, another Board decision introduced the phasing of drawings in order to better enforce conditionality by effectively keeping the requesting members on a short-leash. This was followed by the proliferation of performance criteria, making drawings conditional on their attainment. Although the Board decided in 1968 to limit the number of performance criteria after developing countries argued that the minimum conditionality applied to the drawing by the United Kingdom in 1967 should become the norm, in practice there was no easing of conditionality, particularly after it was given legal sanction in 1969 through an amendment of the Articles of Agreement.12
As a result of these changes in the conditions governing the use of Fund resources, automatic drawing has been confined to the gold tranche with higher tranches bringing tighter conditionality. But since the IMF quotas have lagged considerably behind the growth of world trade, this has effectively meant that countries’ access to balance of payments financing has come increasingly under the policy surveillance of the Fund.
But perhaps the biggest divergence from the Bretton Woods objectives has occurred not with respect to the principle but the content of conditionality. Indeed, while there is now no disagreement on the broad principle of conditionality, policies required for eligibility to drawing from the Fund continue to be the most contentious issue. Through
12 After the 1969 amendment Article V, Sec. 3(d) stated that the “Fund shall examine a request for a purchase to determine whether the proposed purchase would be consistent with the provisions of this Agreement and the policies adopted under them, provided that requests for reserve tranche purchases shall not be subject to challenge”. For the recent trend towards governance-related conditionalities see Kapur and Web (2000), and for the new guidelines Buira (2003).
21 conditionality, the Fund has effectively sought to impose exactly the kind of policies that the postwar planners tried to avoid in countries facing payments difficulties- namely, austerity and destabilizing currency adjustments. In fact, with the breakdown of the Bretton Woods system, currency stability and maintenance of full employment ceased to be the key components of international economic relations (Triffin 1976; UNCTAD 1987).
Austerity has been promoted in the IMF programmes not only when payments difficulties were due to excessive domestic spending or distortions in the domestic price structure, but also when they resulted from external disturbances such as adverse terms of trade movements, hikes in international interest rates or trade measures introduced against the country concerned. Furthermore, the distinction between temporary and structural disequilibria has in effect disappeared, and the Fund has started to operate on the premise that a developing country should interpret every positive shock as temporary and hence refrain from using it as an opportunity for expansion, and every negative shock as permanent, thus adjusting to it by altering the domestic price structure and lowering growth.
The evolution of IMF conditionality has been shaped by shifts in economic and political conditions and interests of its major shareholders. Initially the United States had insisted on some form of conditionality to stem excessive reliance on dollar credits as it was expected to be the only net creditor country to the system particularly after the collapse of the convertibility of the pound sterling. Subsequently, it used conditionality to pursue its national interests, both political and economic. Europe, notably the United Kingdom, initially resisted conditionality because of its need to draw on the Fund’s resources. Subsequently, when they no longer relied on the Fund, conditionality ceased to be a problem for the European countries, including the smaller ones which lost some monetary autonomy in the European Monetary System vis-à-vis Germany but gained considerable
22 protection from Fund conditionality (Akyüz and Flassbeck 2002, p. 98).
The breakdown of the Bretton Woods exchange rate system together with the graduation of the European countries from the Fund pushed the IMF closer to developing countries, both as a crisis manager and as a conduit for structural change. This shift has been accompanied by the introduction of new facilities, not only to address short-term payments difficulties but also structural shortcomings, thereby extending its operations to the domain of development, creating duplications with the Bank. It is through such facilities that the Fund started to distinguish among the problems facing different members.
A Compensatory Financing Facility (CFF) was introduced in the early 1960s as a result of a UN initiative which enabled countries facing temporary shortfalls in primary export earnings to additionally draw on the Fund over and above their normal drawing rights, without the performance criteria normally required for upper credit tranches (Dam 1982, pp. 127-128). However, the semi-automaticity enjoyed by members in their access to this facility was effectively removed by a subsequent decision of the Fund, “accommodating it to the high-conditional, upper credit tranches” (Dell 1985, p. 245). While the CFF has survived alongside standby agreements as a short-term balance of payments facility, a number of other similar ad hoc facilities introduced in the past have been discontinued, including the buffer stock financing facility introduced in the late 1960s and two oil facilities in the 1970s. Of these the latter constituted exceptional steps in IMF lending practices: they had been introduced as deliberate counter-cyclical devices to prevent oil price hikes from triggering a global recession and they allowed the kind of semi-automaticity of drawings advocated by Keynes during the Bretton Woods negotiations (Dell 1986, p. 1207).
Starting in the mid-1970s the Fund’s attention turned increasingly to development
23 issues. In this respect the creation of the Extended Facility in 1974 marks a turning point. It was established as a non-concessional lending facility to address persistent and structural balance of payments problems.13 This was followed by the Structural Adjustment Facility and the Enhanced Structural Adjustment Facility designed to provide concessional lending to low-income countries for structural change. As a result of increased emphasis on poverty reduction, the latter was replaced by a Poverty Reduction and Growth Facility (PRGF) in 1999 as a concessional window for low-income countries.
Perhaps an even more important shift in the role of the Fund in multilateral lending is that it has become a crisis lender and manager for emerging markets. This role effectively started with the outbreak of the debt crisis in the early 1980s when “many developing countries borrowed heavily from multilateral sources in order to finance debt servicing to private creditors” (Sachs 1998, p. 53). Crisis lending has effectively become the dominant financial activity of the Fund after recurrent crises in emerging markets in the 1990s. Two facilities have been established for this purpose (Akyüz 2002b, pp.136-137). The Supplemental Reserve Facility, approved by the IMF’s Executive Board in response to the deepening of the East Asian crisis in December 1997, was designed to provide financing above normal access limits to countries experiencing exceptional payments difficulties, notably in servicing their external debt to private creditors and maintaining capital account convertibility, under a highly conditional stand-by or Extended Arrangement. A second facility, the Contingent Credit Line (CCL), created in Spring 1999 with a view to providing a precautionary line of defence for countries facing the threat of financial contagion, was discontinued in November 2003 because no country had applied for it for fear that it would have the effect of a tocsin in international financial markets, thus stifling their access to credit (Akyüz and Cornford 2002, p. 135; Goldstein 2000, pp. 12-13; and IMF 2003b). As of March 2004, almost two-thirds of total outstanding IMF
24 credits, including drawings from the GRA and PRGF, was accounted for by crisis lending to Argentina, Brazil and Turkey under stand-by agreements (IMF 2004).
B. Development finance
The Bank had been created primarily to address the financing of reconstruction in the war-torn European countries, but this task was effectively taken up by the Marshall Plan as it became clear that it went well beyond the Bank’s lending capacity. Indeed, soon after its inauguration, the Bank started to encounter serious difficulties in meeting requests for loans, mostly from European countries. It had to rely on the United States alone both in lending and borrowing since all loan requests were made in the dollar and there were little prospects for borrowing outside Wall Street. As the Marshall Plan took over European reconstruction, the IBRD portfolio shifted constantly from construction to development loans, and from Europe to developing countries. By the early 1960s less than 10 per cent of total Bank loans were allocated to European reconstruction, with the rest going to development loans. Two-thirds of total IBRD loans, including both for reconstruction and development, went to developing countries and one-third to Europe and Australia.14
As the Bank turned increasingly towards developing countries, the difference between reconstruction and development financing became all too obvious. In particular, most projects in the poorer developing countries did not prove to be viable at moderately below-market rates applied to IBRD loans. An option would have been to extend the
13 See Dam (1982, p.284). For the implications of this mission creep for Bank-Fund relations see Ahluwalia (1999).
14 Oliver (1975: Appendix B). For an account of the evolution of the IBRD in its formative years see Oliver (1975, chaps. IX and X). For a detailed history of the first twenty five years of the World Bank see Mason and Asher (1973) and for its first half century Kapur, Lewis and Webb (1997).
25 Marshall Plan to poorer countries, but this was not the course taken. Instead, the International Development Association (IDA) was established in 1960 with contributions from industrial countries, to be administered by the Bank, in order to provide very long term concessional credits to the poorest developing countries at exceptionally low costs. This was an important departure from the original design which had been premised on the assumption that multilateral financing would take the form of loans rather than gifts or grants and that all borrowers would be charged the same interest rate. It was a clear recognition that a development agency could not simply operate as a bank, extending loans on market terms based on creditworthiness.
While contributions by donors (which now include some middle-income developing countries) through arduously negotiated periodic replenishments have provided the main funding for IDA, the Bank has also used its surplus to supplement financing for IDA countries. The main source of this surplus has so far been the income earned on its paid-in capital plus accumulated reserves rather than spreads on loans of borrowed funds (Gilbert, Powell and Vines 1999, pp. 604-605).
Although during the BWC there was considerable interest in loans to the private sector without government guarantees, this was not part of the original mandate of the IBRD. A possible reason is that postwar reconstruction was expected to involve public rather than private agencies. The shift of emphasis to development finance revived the interest in lending to the private sector, leading to the creation of the International Financial Corporation (IFC) in 1956 to finance private sector ventures in developing countries in partnership with private investors without government guarantees. The increased emphasis on the role of the private sector in development also gave rise to the addition of another institution in 1988, Multilateral Investment Guarantee Agency (MIGA), in order to facilitate foreign direct investment in developing countries by providing guarantees against
26 noncommercial risks to investors. While the demand for IFC credits and MIGA guarantees has increased significantly in recent years, the activities of these institutions remain modest compared to the two main arms of the World Bank Group, IBRD and IDA. Despite the priority given to private sector investment in Article I, direct lending rather than guaranteeing private capital has remained the principal activity of the World Bank.
An equally significant development in the postwar evolution of multilateral lending was the creation of regional development banks (RDBs) and a number of other international organizations to finance development.15 Already in 1940 there was an agreement among a number of Latin American countries and the United States over a convention for the establishment of an Inter-American Bank in order to provide, inter alia, financing for Latin American industrial development. While the convention was never implemented, it was seen as a precursor to subsequent plans for a World Bank.16 The first RDB established was the European Investment Bank, followed a year later in 1959 by the Inter-American Development Bank, by the African Development Bank in 1963 and by the Asian Development Bank in 1965. The main argument advanced in support of RDBs was that they would have a better understanding of the problems and priorities of the countries in their regions and a better dialogue with policy makers, and these would improve the effectiveness of their lending. While it is customary to see RDBs as complementary to the World Bank, there have been instances of tension and a certain degree of competition.17
15 For a detailed account see Kapur and Webb (1994) and Rwegasira and Kifle (1994).
16 According to Oliver (1975, p. 99), the plan prepared for the Inter -American Bank by the United States in the late 1930s “was similar in most respects to subsequent United States Treasury plans for an International Bank for Restructuring and Development. In a sense, the Inter-American Bank plan was the first draft of subsequent plans for a Stabilization Fund and a World Bank”.
17 Kapur and Webb (1994, pp. 232-233) argue that” the Inter-American Development Bank (IDB) was to some extent made possible by the success of the World Bank, but...its establishment was in part a reaction to the political discomfort created by the Bank’s restrictive lending policies”, concurring with Mason and Asher (1973, p. 578) that the creation of the IDB was “at once a tribute and a rebuke to the World Bank”.
27 All RDBs came into being with their own soft windows for concessional lending and mandates to lend to the private sector without government guarantees. Their operations have focussed on project lending although they have also participated in recent years in adjustment programmes supported by the World Bank. Their lending has expanded more rapidly than the World Bank; their share in total lending by multilateral development banks (MDBs) rose from around a quarter in the 1970s to almost 40 per cent in recent years.
The IBRD lending policies underwent a considerable transformation after its inception. As the Bank went deeper into development financing, shortcomings in the original modalities of lending were laid bare. In its early years the Bank had adopted a highly conservative stance, lending at rates close to the market and insisting on high project standards and financial viability of individual projects, and directing its loans to high- yielding infrastructure investment in energy and transportation. This was due, in part, to an effort to establish the Bank’s own creditworthiness in financial markets, and in part to the emphasis placed on the role of capital accumulation in development. It was also believed that infrastructure projects would help attract private investment into industry. As the Bank became deeply engaged in the complexities of development and its emphasis shifted to concessional lending, the importance of development externalities was recognized in the allocation of finance, and its lending started to move towards social sectors and agriculture, a tendency that became particularly pronounced in the 1970s.
Furthermore, as it became increasingly clear that viability of individual projects depended on the functioning of the economy as a whole, the Bank operations started to focus on overall economic conditions and development priorities, moving away from project lending to policy-based lending. This shift culminated in the introduction of ambitious comprehensive market-friendly adjustment programmes with the onset of the
28 debt crisis in the 1980s, supported by structural and sectoral adjustment loans (SALs and SECALs). As it came under increased criticism from civil society for the neglect of key social and environmental concerns, the Bank has constantly added new tasks to its mandate over the past two decades. This mission creep has been accompanied by a proliferation of lending instruments, including special SALs for countries facing financial crisis, and emergency recovery loans for countries emerging from wars, civil disturbances or natural disasters.18
The expansion of the Bank’s operations has been accompanied by the ratcheting of conditionality in very much the same way as in IMF lending. While initially Bank lending was accompanied by project-level or sector-level conditionalities, under structural adjustment programmes they were extended to the macro level. This has encouraged a high degree of collaboration between the two BWIs, and often the Bank policy conditionality drew upon and complemented the policy advice and performance criteria developed by the Fund, and rarely has an adjustment loan been given without the IMF’s consent.19
The increase in the influence of the Bank in key economic decisions in borrowing countries has not been matched by its contribution to development finance. In real terms in the past two decades there has barely been any increase in net flows to developing countries from the Bank, including both IBRD loans and IDA credits, because of declines in net flows from IBRD (table 1). Even if the sharp decline in 2002 is excluded, the increase in average net flows during 2000-2001 compared to 1980 was roughly equal to the
18 On various lending instruments see World Bank (2001). For an account of mission creep by a former Managing Director of the World Bank, see Einhorn (2001).
19 The most notable exception is the Argentinian adjustment programme in 1988 when the Bank went ahead with lending even though the negotiations between Argentina and the IMF on an EFF had broken down: see Ahluwalia (1999, p. 5).
29 increase in import prices of developing countries so that in terms of purchasing power over imports, net flows from the WB were broadly unchanged. More importantly, net transfers (that is, net flows minus interest payments) from the IBRD have been negative in almost every year since 1990, for both low-income and middle-income countries, while net transfers on IDA credits have kept up. Thus, the IBRD has not made any contribution to the financial needs of developing countries as a whole in recent years, other than providing finance to service its outstanding claims, and the main contribution of the Bank to development finance has taken the form of highly concessional IDA credits. Ironically, the “Bank had strenuously objected to soft lending from the moment the idea was first mooted”(Kapur and Webb 1994, p. 231), but such lending has in effect become its principal contribution to development finance.20
(Table 1 here)
IV. International financial markets and multilateral lending
The past sixty years have thus seen a gradual shift in the objectives and modalities of multilateral lending. In fact, in this respect neither the Fund nor the IBRD has been rendering the functions they were originally designed to perform. The principal activity of the Fund is no longer lending to countries facing temporary difficulties in their current account payments, but financial bailout operations associated with capital account crises and, to a lesser extent, provision of concessional development finance to low-income countries. Similarly, the main contribution of the World Bank to development finance is concessional IDA credits rather than IBRD loans.
20 It has also been noted that IDA “helped finance repayments to the ... IBRD while also mitigating pressures on the IBRD to make loans to countries with low creditworthiness” (Kapur 2003, p. 2). This implies that in the absence of
30 The main issue, however, is not simply whether the original rationale for multilateral lending is still valid today, but the role that multilateral lending can be expected to play in attaining the internationally agreed goals, and whether current modalities of multilateral lending can meet these tasks. This depends very much on a correct diagnosis of the nature and effects of private capital flows to developing countries. In this respect recent history provides considerable insight.
The postwar era has seen two boom-bust cycles in private capital flows to developing countries: the first beginning in the early 1970s and ending with the debt crisis, and the second beginning in the early 1990s and ending with the East Asian and Russian crises.21 These two cycles generated similar amounts of cumulative net private inflows in constant (2000) dollars: some $1.150 billion in the first boom and $ 1.240 in the second. The first boom was driven by the rapid expansion of international liquidity associated with oil surpluses and growing United States external deficits, and facilitated by financial deregulation in industrialized countries and rapid growth of Eurodollar markets. Excess liquidity was recycled in the form of syndicated bank credits, and this was encouraged by the BWIs for fear of a collapse of global demand. However, with increased debt servicing difficulties brought about by the hike in United States interest rates and global recession, there was a sharp cutback in bank lending, forcing debtor countries to generate trade surpluses through cuts in imports and growth.
The second boom came after almost ten years of interruption in private lending to
IDA, net transfers on IBRD loans could have been greater.
21 Boom-bust cycles characterise not only the postwar experience, but almost the entire history of private capital flows to developing countries. The boom in private flows to Latin American countries that started soon after their independence around 1820 was followed by widespread defaults after ten years, and disappearance of international liquidity to the region until around 1850. Again the boom of the 1920s was followed by widespread defaults and cutbacks in private lending in the 1930s. For a more detailed account of these cycles see UNCTAD (2003, chap. II, and pp. 129-132), UNCTAD (1998, Part One, chap. III) and Kregel (2004).
31 developing countries. It was encouraged by the success of the Brady Plan, and liberalization, privatization and stabilization in developing countries, as well as by a rapid expansion of liquidity in the United States and Japan in conditions of stagnation and recession. Unlike the first boom, a large proportion of private inflows were in equity investment, rather than international lending. In most cases they were driven by prospects of quick capital gains and short-term arbitrage opportunities. Initially capital inflows often exceeded the current account needs of recipient countries, resulting in currency appreciations and thereby creating a need for themselves. When they were reversed, many debtor countries were again faced with negative net transfers, and loss of income and employment.
A feature of these cycles is that they were driven by temporary and special factors, including monetary policies and financial policies in industrial countries, rather than long- term fundamentals in the recipient countries. As a result, aggregate flows to developing countries manifested a degree of instability not justified by changes in the underlying fundamentals in the recipient countries. For instance in the last cycle, total annual net private capital inflows fell from more than $200 billion in 1996 to less than $20 billion in 2000-2001, and after the East Asian crisis until 2002 net international private lending to developing countries was negative. These swings in the volume of private capital are also mirrored in sharp changes in their terms and conditions. As noted, booms tend to be associated with the underestimation of risks and relatively low spreads, while crises often lead to overpricing of risks, generalized increases in spreads and shortening of maturities.22 The increased instability of private flows explains why many developing countries today accumulate excess reserves at the expense of imports and growth, and at high carry costs,
22 See UNCTAD (2003, p. 27, Figure 2.1) and Cunningham, Dixon and Hayes (2001, p. 179) for the behaviour of spreads after the East Asian and Russian crises. For most emerging markets spreads were highly clustered and exceptionally low before the Asian crisis, but subsequently their dispersion increased sharply and they became “excessively high” for several emerging markets (Sy 2001).
32 even though the greater access of these countries to international financial markets should, in principle, be expected to reduce their needs for reserves (UNCTAD 1999, pp.108-109).
Although the terms and conditions of private loans vary significantly through boom- bust cycles, on average their maturities are short and costs are high: that is, they are not available at “reasonable terms”. Over the past ten years, the average maturity of private lending to middle-income countries was around half of the average maturity of official lending (World Bank 2002). Again, the average spread of an emerging-market bond index constructed by the World Bank (2000, p. 23, chart 1.13) exceeded 700 basis points during the 1990s and never fell below 400 basis points.
This is not to say that all emerging markets face similar conditions or the underlying fundamentals and domestic policies do not matter. Indeed after the debt crisis of the 1980s some developing countries, notably in East Asia, continued to enjoy access to international capital markets despite the slowdown in aggregate lending to developing countries. Again, during the recent bouts of instability, contagion has been partial and markets have had a tendency to differentiate among different classes of sovereign borrowers with respect to risks (Cunningham, Dixon and Hayes 2001, p. 177; and UNCTAD 2003, p. 30). However, even when differentiating across borrowers, markets tend to overreact and behave pro- cyclically, withdrawing capital almost totally from countries perceived to have high risks, and generating fragility and vulnerability in the countries they favour. This contrast is once again visible in the recent behaviour of private capital flows. With the decline in interest rates and expansion of international liquidity, private capital has started to flow into some East Asian economies with strong payments positions, raising the risk of unsustainable bubbles (Pesek 2003; Crampton 2004; and Williams 2004), while many other middle- income countries with weak payments positions and in need of external finance have been unable to receive adequate finance because of unfavourable market perceptions of their risk
33 profiles.
The fundamental issue here is to what extent, in a world of inherently unstable capital flows, emerging markets can secure reasonably steady and adequate flows of private capital by their own policy action. Certainly, this calls for a policy of strategic integration, as opposed to a full and rapid integration, into global capital markets. The space to avoid boom-bust cycles and unsustainable financial bubbles through a judicious combination of macroeconomic, exchange rate and regulatory policies, including capital account regulations, is certainly wider than has so far been exploited by policy makers in developing countries (Akyüz 2004). In particular, several East Asian countries with sound fiscal and balance of payments positions, high levels of domestic savings and investment, competitive export industries and high growth potentials can, in principle, tap international capital markets and mobilize adequate private financing without going through bubbles and boom-bust cycles. Indeed, many of the more successful economies of that region, including Korea and Taiwan, resorted to extensive borrowing from international capital markets in their early phases of industrialization without experiencing the kind of instability that devastated the region during 1997-1998, thanks to effective regulation and control over such borrowing. Certainly, under a policy of strategic integration, the need of more successful countries for multilateral finance could be limited to occasional IMF lending in the form of counter-cyclical provision of international liquidity for current account financing.
But most other developing countries cannot attract and sustain adequate amounts of private finance at reasonable terms because of their deep seated structural and institutional weaknesses. This includes not only low-income but also middle-income countries. Despite improved perceptions about emerging market risks and substantial compression of spreads in recent months (IIF 2004), less than half of the 29 emerging-markets now enjoy
34 investment-grade status, while the large majority of middle-income countries are rated as speculative and face very high spreads. This latter group includes many middle-income Latin American countries with chronic savings and foreign exchange gaps, weak industrial export bases and excessive debt. Their high degree of dependence on external capital renders them extremely vulnerable to trade and financial shocks. Such shocks pushed them into a downward spiral in the 1980s following extensive borrowing from international banks during the previous decade. But policies introduced in response to the debt crisis with the support of the BWIs have left many of these countries in conditions as fragile as those prevailing in the 1980s (UNCTAD 2003, chap. VI).
The dilemma is that it is precisely those countries that do not have much need for foreign capital that can attract and sustain relatively large amounts of private capital at “reasonable terms”. By contrast, countries with large structural savings and foreign exchange gaps, and high levels of debt need substantial and sustained inflows of foreign capital, but they cannot rely on private flows without running into recurrent crises and stop- go development. A combination of borrowing at high interest rates and weak growth fundamentals is indeed a recipe for disaster. For a debtor country to enjoy a net resource transfer from abroad to supplement its domestic savings and raise the level of investment, the rate of increase of its foreign borrowing should exceed the rate of interest on its outstanding debt. Since the capacity of the country to service its debt is effectively determined by income growth, debt would not be sustainable when it is growing faster than income. This would be the case whenever the rate of interest on outstanding debt exceeds the potential growth rate. In such a case, external borrowing becomes a Ponzi process whereby servicing existing debt will depend on the willingness of the creditors to lend indefinitely. External shocks leading to higher interest rates (due to rising spreads brought about by a shift in market perceptions of risks or increases in international rates) or lower growth can translate debt accumulation into an unsustainable process. Since reverse
35 transfers necessitated by cuts in lending cannot be maintained indefinitely, defaults become inevitable.23
Since many of these countries, particularly those at relatively high levels of per capita income, cannot expect to fill their savings and foreign exchange gaps indefinitely with multilateral finance at favourable terms, the only way for such countries to get out of this impasse is to increase self reliance and improve economic performance. On the other hand, since in most such cases, the “ever-present possibility that the debt might not be serviced is the single most important reason for the excessively high international risk premium charged on most sovereign borrowing”, a “viable exit from this vicious circle of low investment and growth, high interest rates and rising indebtedness” calls for the elimination of unsustainable debt in an orderly way (UNCTAD 2003, p. 146).
V. Multilateral lending today: the way forward
A. IMF lending
While there is a consensus on the role of the IMF in the provision of short-term international liquidity, there is little agreement on the objectives to be pursued by such lending. Most major creditor governments continue to resist proposals for the enlargement of IMF quotas and SDR allocation for increased provision of international liquidity to developing countries facing export shortfalls or rising import bills,24 but they have been willing to support Fund lending, in excess of official limits, to countries facing capital
23 For an analysis along these lines, drawing on the Minskian endogenous fragility, see Kregel (2004).
24 For the debate on the provision of international liquidity during the debt crisis see Akyüz and Dell (1987, pp. 44- 48).
36 account crises for fear of their adverse repercussions for international creditors and investors. Such assistance has often come after the collapse of the currency and hikes in interest rates in the form of bailout operations designed to keep countries current on their debt repayments to private creditors, to maintain capital account convertibility, and to prevent default.
The main problem with such lending is that not only it fails to protect countries against adverse consequences of financial shocks, but it also tends to aggravate, rather than rectify, market failures by creating creditor moral hazard. It prevents market discipline and encourages imprudent lending since private creditors are not made to bear the consequences of the risks they take.25 Furthermore, even when the external debt is owed by the private sector, the burden ultimately falls on taxpayers, because governments in debtor countries are often obliged to serve as guarantors.
Recent years have seen increased calls for the IMF to become an international lender of last resort for emerging markets not only to resolve but also to pre-empt crises (see, e.g. Fischer 1999). The main idea is that if IMF stands ready to provide liquidity to countries with sound policies, these countries would be protected from contagion and financial panic so that a lender of last resort facility would have a preventive role. Clearly, such a step would involve a fundamental departure from the underlying premises of the Bretton Woods system. This is particularly notable in the report of the Meltzer Commission (2000), which virtually proposes, inter alia, the elimination of all other forms of IMF lending, including those for current account financing which should, in their view, be provided by private markets. As noted by a former United States Treasury Secretary,
25 For a survey of empirical evidence on the effect of IMF intervention on debtor and creditor incentives see Haldane and Scheibe (2003) who “find concrete evidence of creditor-side moral hazard associated with IMF bail-outs” (p. 1). See also Mina and Martinez-Vazquez (2002) who conclude that IMF lending generates moral hazard in international financial markets from the perspective of the maturity composition of foreign debt.
37 such a shift in the nature of IMF lending, from current account to capital account financing, would imply that only a small number of relatively prosperous emerging economies would be eligible for IMF financing (Summers 2000a; and Summers 2000b, p.14).
Furthermore, there are a number of difficulties in translating the IMF into a genuine international lender of last resort, and proposed arrangements could compound rather than resolve certain problems encountered in the current practice of IMF bailouts.26 The effective functioning of such a lender would require discretion to create its own liquidity (or to have unconstrained access to international liquidity). While in principle various arrangements are possible (such as a new role for the SDR or allowing the Fund to go to markets), they are all problematic (see Akyüz and Cornford 2002, p. 133-134). More importantly, the terms of access to such a facility pose additional problems. The conditions of lender-of-last-resort financing, namely lending in unlimited amounts and without conditions except for penalty rates, would require much tightened global supervision over borrowers to ensure their solvency, but this would encounter not only technical but also political difficulties. Pre-qualification, that is allowing countries meeting certain ex ante conditions to be eligible to lender-of-last-resort financing, as in the case of ill-fated CCL, involves several problems. First, IMF would have to act like a credit-rating agency. Second, it would be necessary constantly to monitor the fulfilment of the terms of the financing to ensure that the pressures on the capital account of a qualifying country have resulted from a sudden loss of confidence amongst investors triggered largely by external factors rather than macroeconomic and financial mismanagement. In these respects difficulties are likely to emerge in relations between the Fund and the member concerned. Last but not least, creation of a lender of last resort may aggravate rather than eliminate
26 The dissenting members of the Meltzer Commission pointed out that the most damaging proposals relate to the IMF’s role in financial crises (Fidler 2000); see also Eichengreen and Portes (2000) and Wolf (2000). In this respect the Commission Report is not consistent; as pointed out by DeLong (2000, p. 2), while it assigns a lender of last resort role to the Fund for lending solvent but illiquid governments, it condemns the Fund for its loans to Mexico in 1995 and
38 creditor moral hazard and sound policies cannot always succeed in preventing imprudent lending and speculative capital flows.
Difficulties associated with IMF crisis lending have given rise to various proposals to involve the private sector in the resolution of emerging-market crises and limit IMF bailout operations.27 A proposal made by the UNCTAD Secretariat already during the debt crisis of the 1980s is to build on national insolvency procedures, notably Chapter 11 of the United States bankruptcy code, in order to prevent financial meltdown and economic crises in countries facing difficulties in servicing their external obligations, and to involve the private sector in the resolution of crises.28 Attaining these objectives would not require fully-fledged international bankruptcy procedures, but the application of a few key principles:
$ A temporary debt standstill accompanied by exchange controls in order to stop speculative attacks and prevent draining of reserves.
$ Provision of debtor-in-possession financing, automatically granting seniority
recommends against any increase in the IMF’s resources. See Meltzer (2001) for his comments on the critics.
27 The list of institutions and experts who put forward various proposals for mechanisms to overcome moral hazard and involve the private sector in the resolution of financial crises includes the Group of 22 (1998), the Council of Foreign Relations Independent Task Force (CFRTF 1999); the Emerging Markets Eminent Persons Group (EMEPG 2001); and the High-Level Panel on Financing for Development (Zedillo, 2001). For a discussion of the IMF’s crisis lending and various proposals for reform see Goldstein (2000), Haldane (1999), Akyüz (2002b) and Eichengreen (1999 and 2002).
28 See UNCTAD (1986, annex to chap. VI). The need for orderly workouts was subsequently raised by Sachs (1995), and the proposal was revisited by the UNCTAD Secretariat during the East Asian crisis (UNCTAD 1998, pp. 89-93). For a further discussion see Radelet (1999) and Akyüz (2002b). The idea of establishing orderly workout procedures for international debt goes back even further. In 1942, in a report by the United States Council on Foreign Relations attention was drawn to “interwar disputes between debtors and creditors and even between creditor governments,...and it was suggested that (T)here is need for exploration of the possibilities of minimizing the source of international conflict through the establishment of ... a supranational judicial or arbitral institution for the settlements of disputes between debtors and creditors” (Oliver, 1971, p. 20).
39 status to debt contracted after the imposition of the standstill.
$ Debt restructuring including rollover and write-off, based on negotiations between debtors and creditors, and facilitated by the introduction of automatic rollover and collective action clauses in debt contracts.
In view of the failure of IMF interventions to stabilize emerging-market economies facing financial difficulties, the IMF Board finally recognized that it may indeed be necessary to impose a unilateral temporary debt standstill,29 and the IMF Secretariat developed a proposal along the above lines.30 However, the original proposal has been significantly revised and, inter alia, the provision for statutory protection to debtors in the form of a stay on litigation removed because of the opposition from Wall Street and the United States.31 Governments in some debtor countries, notably in Latin America, have also been reluctant to support the proposal for fear of impairing their access to international capital markets. Even the watered down version of the IMF’s Sovereign Debt Restructuring Mechanism (SDRM) proposal has not enjoyed a consensus and has been pushed to the back of the agenda. However, despite these difficulties, this initiative reflects a growing recognition that the approach so far adopted, built on the principle of maintenance of free capital flows and guaranteed repayment to creditors, and supported by hefty rescue packages, is not the best way to intervene in emerging-market crises.
29 See IMF (2000). For further discussion of the debate in the IMF see Akyüz (2002b, pp. 123-128).
30 See the statement by its First Deputy Managing Director, Krueger (2001). There were some differences between the UNCTAD proposal and the original IMF proposal. In the UNCTAD proposal the decision to impose standstill should rest with the debtor country but would then be subject to an examination and endorsement of an independent panel very much along the lines of the WTO safeguards procedures. In the IMF proposal, by contrast, the “standstill would be activated if a request by the debtor country was endorsed by the Fund” (Krueger 2001, p. 9).
31 See IMF (2003a) for a brief description of the SDRM and the background information.
40 Given the persistent global economic instability and pro-cyclical behaviour of international financial markets, a reform of IMF lending should pay greater attention to the provision of counter-cyclical financing in response to external trade and financial shocks. While it should be recognized that money is fungible and in practice it is not always possible to identify the need catered for by a particular loan, it is important to ensure that Fund resources be used to finance current account rather than capital account transactions.
First, IMF lending to counter sharp declines in private capital flows should aim at maintaining imports and other vital current account transactions, and the level of economic activity, rather than debt repayment to private creditors and capital account convertibility, which should be dealt with primarily by temporary debt standstills and exchange controls. Such lending should be available both to countries experiencing crises as well as those threatened by contagion. This implies that there should be strict limits to IMF crisis lending since otherwise it would be difficult to ensure private sector involvement. Such an approach has been supported in a report to the Council on Foreign Relations which argued that the IMF should adhere consistently to normal access limits and that only “in the unusual case in which there appears to be a systemic crisis (that is a multicountry crisis where failure to intervene threatens the performance of the world economy and where there is widespread failure in the ability of private capital markets to distinguish creditworthy from less creditworthy borrowers), the IMF would return to its ‘systemic’ backup facilities” (CFRTF 1999, p. 63). While the concerns underlying different lending policies for systemic and non-systemic crises may be justified, in practice exceptions to normal lending limits could leave considerable room for large-scale bailout operations and excessive IMF discretion. Such a route should be taken only with the participation of private lenders in crisis resolution. It is also important to separate multilateral lending from bilateral arrangements in order to reduce the scope for undue influence over Fund policies by some of its major shareholders. Finally, greater delineation of Bank-Fund activities
41 requires that crisis lending should be the sole responsibility of the Fund, and the Bank should stay out of provision of short-term liquidity.32
The second area of IMF lending is counter-cyclical support for countries facing temporary payments difficulties in their current accounts due to shortfalls in their export earnings, surges in their import prices or hikes in interest rates on their external debt. Such lending should be available not only to poorer developing countries that have little or no access to international liquidity, but also to emerging markets which normally have access to private finance but often see their access impaired at times of current account difficulties because of pro-cyclical behaviour of markets. Such lending is in line with the original tasks assigned to the Fund, and should be designed in the same way as the original CFF.
Finally consideration should be given to the creation of a global counter-cyclical facility to be used at times of a sharp downturn in world economic growth and international trade. As noted above, suggestions were made for the use of Bank loans for this purpose during the Bretton Woods negotiations, but such a task is better suited for the IMF. Indeed, the Fund performed this task during the 1970s through the two oil facilities discussed above. Again, in access to such a facility countries should enjoy the kind of semi- automaticity advocated by Keynes.
The exact modalities that IMF lending should take in these areas will go beyond the scope of this paper. However, the fundamental point here is that the IMF should limit itself to short-term, counter-cyclical lending for current-account purposes, and there is no sound
32 This view is also held by the majority in the Meltzer Commission (2000, p. 11). See also Gilbert, Powell and Vines (1999, p. 622) who note that during the 1998 crisis “the Bank provided around $8 billion in short-term liquidity in the packages of lending to Thailand, Indonesia and Korea. This ‘defensive’ lending had little to do with promoting development (in the normal sense of that expression). If it were to be repeated, such emergency lending could
42 rationale for the Fund to engage in other forms of lending, including capital account financing and development finance. As noted, there is a wide concern about the problems posed by IMF bailout operations and a number of proposals have been made to eliminate or reduce the need for such lending. On the other hand, the view that the Fund should no longer be involved in development is now broadly shared (see e.g. Meltzer Commission 2000, p. 4; and Stiglitz 2002b, p. 232).
A restructuring of IMF lending along these lines implies a radical reform since, as noted, much of the recent activities of the Fund have focused on financial bailout operations and concessional lending for development. Such a reform should also address the issue of adequacy of IMF resources. While normal access limits of 100 per cent of quota annually and 300 per cent on a cumulative basis should in principle be observed in IMF lending, it should also be recognized that IMF quotas have lagged behind the growth of global output, trade and financial flows. According to one estimate, adjusting quotas for the growth in world output and trade since 1945 would require them to be raised by three and nine times respectively (Fischer 1999). A large increase in the IMF quotas would also widen the gold tranche, and help restore the balance between low-conditionality and high- conditionality loans.
B. Development finance
As noted above, there was considerable ambiguity about the role that the IBRD was expected to play vis-à-vis private capital markets and investment during the Bretton Woods negotiations. It should thus not come as a surprise that questions are raised regarding the rationale of the Bank continuing and even expanding its lending despite the rapid
severely destabilise the Bank’s normal development lending”.
43 development and closer integration of international capital markets. Indeed, with the recovery of private capital flows to developing countries in the early 1990s, it has been increasingly argued that the original rationale of multilateral lending is no longer valid. However, this argument has variants with different implications for the operation of MDBs and their borrowers. On one view, there are other compelling reasons for such lending to continue, while on another, multilateral lending to countries with access to international capital markets should be discontinued.
An argument deployed in favour of continued multilateral development lending is that MDBs are superior to private agents in generating and disseminating knowledge and promoting sound policies, and policy advice and technical assistance should be bundled together with multilateral lending in order to avoid incentive and agency problems. It is held that since risk-return profiles of projects depend on the overall policy environment, improvements in policy making resulting from conditionality attached to multilateral lending provides a rationale for such lending even in the absence of other types of market failure. Thus, “it is more plausible to locate the Bank’s comparative advantage in assisting developing countries in the presence of weaknesses and distortions in member countries’ domestic political processes than in overcoming the international capital market imperfections” (Gavin and Rodrik 1995, p. 311).
On this view, developed further in Rodrik (1995), compared to private investors, these institutions have considerable know-how and expertise in development policy resulting from their research and practical experience, and access to a significant amount of information on the overall policy environment in developing countries. They are also better positioned to get good policies implemented through conditionality attached to their lending. These allow them to finance projects that private lenders find too risky to undertake. By doing so they also help mobilize private capital by improving the risk-return
44 profiles of private investment. It is also argued that such a task cannot be accomplished with multilateral agencies monitoring government policies on behalf of bilateral and private lenders since there may be severe incentive problems in the provision of these functions in the absence of direct lending.33
Such tasks would constitute a sound rationale for multilateral lending provided that they are properly fulfilled. But this is not what the record shows. The evidence does not support the catalytic role of multilateral lending.34 More importantly, serious questions have been raised by a wide range of researchers, both inside and outside the Bank, about the effectiveness of conditionality and the success of projects financed by multilateral lending.35 Indeed, it has been increasingly argued that ex ante conditionality has not been successful in preventing policy failures, and it would need to be replaced by a new mechanism of ex post conditionality, rewarding rather than trying to buy sound policies. Some of these authors would like to see the World Bank become an effective “knowledge bank” focusing on the provision of knowledge, policy advice and technical assistance, but they nevertheless argue that IBRD should continue lending to middle-income countries because of complementarity between technical assistance and lending, and because such lending would generate funding for IDA (Gilbert, Powell and Vines 1999; and Gilbert
33 A somewhat different reason based on conditionality was advanced by the former Treasury Secretary of the United States. Commenting on the Meltzer Commission proposals Summers (2000a) argued that an implication of discontinuing lending to middle-income countries would be that “our capacity to influence the direction and policies of these economies in ways that support global interest would be lost”.
34 Rodrik (1995) himself could not find any evidence that multilateral lending encourages private flows: “Non- concessional flows from multilateral sources other than the IMF have a negative effect on subsequent private flows “ (p. 24). His finding that IMF flows tend to follow private flows lends support to arguments that such lending serves to bail out private investors. A study at the Bank also found “a counter-cyclical relation between non-concessional multilateral flows (i.e. multilateral lending to middle-income countries) and private flows” and noted that the beginning of such an inverse relation coincided with “the introduction of adjustment lending” (Ratha 2001, p. 7).
35See e.g. Stiglitz (2002b, chap. 2), Gilbert, Powell and Vines (1999, pp. 616-619), Ocampo (2001, pp. 13-14), Meltzer Commission (2000, pp. 43-62) and Meltzer (2001).
45 2000).
However, these arguments are not very persuasive. There are many organizations in the United Nations system which provide effective technical assistance to developing countries without combining it with lending. More importantly, as noted by the same authors, the IBRD loan-related income has not made a contribution to IDA (Gilbert, Powell and Vines 1999, pp. 604-605). This point was also made by Meltzer (2001, p. 11): “The World Bank and others [claim] that ending loans to middle income countries would harm the poorest countries by reducing the Bank’s income. This claim has no merit. The Bank lends at a rate very little above its own cost of funds; it adds a fractional fee to cover administrative expense. The only ‘profit’ on the loan comes from the allocation of a portion of the Bank’s costless capital. The same capital would be available to support grants. There will be no diminution of resources”.
The desire to increase loan-related income has no doubt played an important part in the recent decision of the Bank to raise its effective margin in IBRD lending. However, this is a source of tension among developing-country members of the Bank. Promoting the already established practice of voluntary contributions by more advanced developing countries would provide a better option than using IBRD lending to supplement IDA.
The majority in the Meltzer Commission took the proposition that middle-income countries now have access to international financial markets to its logical conclusions and recommended a phased graduation of these countries from multilateral lending and a conversion of IDA credits into grants. After proposing that “all resource transfers to countries that enjoy capital market access (as denoted by an investment grade international bond rating) or with per capita income in excess of $4.000 would be phased out over the next five years” (Meltzer Commission 2000, p.10) the report goes on to argue for limits on
46 official assistance for countries above $2500 per capita income. For the remaining 80-90 poorest countries, the Bank should provide grants not credits. Such a reform would indeed imply that these institutions would stop acting as banks; they should in fact be renamed development agencies. Accordingly the IFC should be merged into the World Bank and the MIGA eliminated.
While the principle of graduation is incontrovertible, there are certain difficulties in the way the proposal is formulated. Experience clearly shows that investment grade status is not an attribute inextricably linked to long-term structural fundamentals of developing countries, or strongly correlated with their levels of per capita income. For instance, if current ratings were to be taken as the basis of graduation, China and India would be excluded from IBRD lending while Argentina would continue to enjoy access. Again, as shown by recent history, credit ratings of emerging markets are capable of undergoing sharp and unexpected changes according to the success of countries in managing volatile capital flows, and a favourable rating at any point in time is no guarantee for uninterrupted access to private finance in the future. Currently half of the emerging-market economies rated as investment grade, including many Asian countries and Mexico, were considered as high-risk markets at one time or another during the 1990s while some of the countries such as Argentina which are deep down in the list now were rated very highly. By the same token, it is not clear how many countries currently rated as investment grade will retain this status in the coming years.
In its response to the Meltzer Commission, the United States Treasury also noted the fragility of market access and argued that “even relatively productive emerging markets face severe limitations in the volume of private capital that is reliably available for long- term development investments with the medium to longer-term maturities that are necessary. Moreover, the private capital that is available comes with interest rates that are
47 prohibitive for development programmes” (US Treasury 2000, pp. 7, 27). An argument advanced in support of the Meltzer Commission’s recommendation is that the Bank is able to finance socially useful projects at reasonable terms because it “receives a government guarantee of principal and interest” and “[i]f private lenders received the same guarantee, they would not care what the loan financed (Meltzer, 2001. P. 12). However, this argument concedes that the original rationale is still valid. Furthermore, by pooling risks the Bank reduces the cost of guarantee compared to an alternative scheme where governments provide guarantees for individual private loans to developing countries.
The proposal for translating loans into grants for the poorest countries has enjoyed considerable support, but it has generally been considered impractical because of lack of funds. Thus, it has been argued that “the shift to grant-based funding would drastically reduce the total amount of official resources that can be brought to bear in these economies, and bringing to an end any capacity for concessional flows to be re-lent. It bears emphasis that roughly half of the $20 billion in IDA is made up of ‘reflows’ of funds due to past recipients’ repayment of loans” (Summers 2000b, p. 15). But the new administration in the United States went on to propose that 50 per cent of the resources of IDA be disbursed as grants. This was also initially resisted by the European governments on grounds that it would jeopardise the Bank’s long-term financial viability.36 However, eventually the IDA deputies “recognized the advantages of increasing concessionality ... and endorsed an expansion of IDA grants during the IDA13 period, in the range of 18 to 21 per cent of overall IDA13 resources” (IDA 2002, p. 27).
A conclusion that could be drawn from the recent debate on the rationale of multilateral lending is that the MDBs generally and the World Bank particularly should
36 On different views and exchanges on the financial consequences of converting IDA credits to grants see Lerrick and Meltzer (2001, 2002) and Sanford (2002 and 2003), Salazar (2002) and Birdsall (2002).
48 focus their activities on the poorest developing countries and the provision of global public goods, and should adopt a graduation strategy for the more advanced developing countries. As noted, neither market rating nor per capita income would provide appropriate criteria for graduation. The Bank could scale down its lending activities in a number of more successful middle-income countries which do not have much need for multilateral finance. These countries can graduate and rely on private markets for a relatively stable flow of finance, provided that they pursue a policy of strategic integration into international financial markets to deter speculative capital flows and avoid serious instability and crises. The main contribution of the BWIs to these countries should be in the formulation of such a strategy, but this calls for a major departure from the approach adopted so far to financial liberalization and openness.
For many other middle income countries, continued dependence on multilateral finance is the outcome of misguided development policies. These countries could only be graduated from multilateral lending in the context of rapid and sustained economic growth leading to rising domestic savings and increased and stable flows of long-term private capital.37 Sustaining such a process calls not only for multilateral lending but also a new approach to development policy, learning from the mistakes of the policies based on the Washington Consensus, as well as drawing lessons from more successful development experiences.
For the large majority of poorer developing countries there is no alternative to the provision of multilateral finance at highly favourable terms. For a number of reasons grants are preferable to concessional IDA credits. In particular, many of these countries are already highly indebted and in need of a substantial debt write-off, and there is not
37 Domestic savings and sustained inflows of private capital lag rather than lead growth, but this is not a spontaneous process. For an analysis along these lines see UNCTAD (2000).
49 much to be gained by adding to their stock of debt and interest payments, however concessional they may be. This means that for this group of countries, multilateral financing would need to advance in a direction totally different from that envisaged by the postwar planners. This was already evident in the better days of development cooperation:
“Supposing that the international capital markets would become more nearly perfect over time, John J. McCloy [the second President of the IBRD] thought of the Bank as a temporary institution which would go out of business if it were successful, for it would no longer be needed between productive borrowers and private lenders. What neither White nor McCloy understood was that commercial interest rates and repayment conditions may be ‘unreasonable’ for many less developed countries if they are to develop rapidly or perhaps, at all. But this is to say no more than that economic development has turned out to be a great deal more difficult to induce than anyone in the wealthier nations supposed a quarter century ago. If the probability of widespread development is to be increased, grant-type foreign assistance must be provided on a substantial scale and its direction and use must be wisely managed”. Oliver (1975, p. 259).
How this cooperation is organized would have a major influence on its effectiveness. In this respect current arrangements regarding IDA leave a lot to be desired. It has been persuasively argued by Kapur (2003, pp. 2-5) that the creation of IDA reduced the institutional autonomy of the Bank and subverted its governance by allowing political leverage and interference by its major shareholders, and that translating IDA credits into grants may erode the Bank’s autonomy further by increasing dependence on donor contributions.
A possible way out of this impasse is to establish a development cooperation endowment (DEC) large enough to generate adequate and expanding amounts of grant for the poorest countries, run by a professional secretariat without political interference from its contributors. Currently gross annual disbursements from IDA vary around $6 billion and net disbursements less than $5 billion. An endowment of $150 billion earning about 8 per cent per annum in international capital markets would generate sufficient revenues to
50 cover a similar level of grants as well as reserves needed to allow a growth of some 5 per cent over time. This could be established by voluntary contributions from governments, both in the North and the South, private foundations and wealthy individuals. In this respect, various proposals that have been put forward to mobilize new and independent sources of development finance, ranging from a global lottery to environmental and financial taxes are also worth pursuing.38 Existing IDA resources could become part of such an endowment provided that the donors agree to hand them over to an independent secretariat.
38 The UNU/WIDER project on Innovative Sources of Development Finance contains a number of such proposals. For a synthesis see Atkinson (2003).
51 REFERENCES
Ahluwalia, Montek S. 1999. “The IMF and the World Bank in the New Financial Architecture”. International Monetary and Financial Issues for the 1990s. Vol. XI. Geneva: UNCTAD.
Akyüz, Yilmaz, ed. 2002a. Reforming the Global Financial Architecture. Issues and Proposals. London: Zed Books.
Akyüz, Yilmaz. 2002b. “Crisis Management and Burden Sharing”. In Yilmaz Akyüz, ed. Reforming the Global Financial Architecture. Issues and Proposals. London: Zed Books..
Akyüz, Yilmaz. 2004. Managing Financial Instability and Shocks in a Globalizing World. Tun Ismail Ali Chair Public Lecture. Kuala Lumpur: Bank Negara.
Akyüz, Yilmaz and Sidney Dell. 1987. “Issues in International Monetary Reform”. In International Monetary and Financial Issues for the Developing Countries. Geneva: UNCTAD.
Akyüz, Yilmaz and Heiner Flassbeck. 2002. “Exchange Rate Regimes and the Scope for Regional Cooperation”. In Yilmaz Akyüz ed. Reforming the Global Financial Architecture. Issues and Proposals. London: Zed Books.
Akyüz, Yilmaz and Andrew Cornford. 2002. “Capital flows to developing countries and the reform of the international financial system”. In Deepak Nayyar, ed. Governing Globalization. New York: Oxford University Press.
Atkinson, Anthony B. 2003. Innovative Sources for Development Finance: Over-Arching Issues. WIDER Discussion Paper No. 2003/88.
Birdsall, Nancy. 2002. In Praise of Grants. Financial Times. 26 February.
Bryant, Ralph C. 1980. Money and Monetary Policy in Interdependent Nations. Washington: The Brooking Institutions
Buira, Ariel. 2003. An Analysis of IMF Conditionality. G24 Discussion Paper No. 22. Geneva: UNCTAD.
Cassen, Robert. 1994. Does Aid Work? Oxford: Clarendon Press.
Crampton, Thomas. 2004. “Asia’s rally defies most expectations”. International Herald Tribune (online). 21 January.
52
CFRTF 1999. Safeguarding Prosperity in a Global Financial System. The Future International Financial Architecture. Washington, DC, Council on Foreign Relations Independent Task Force, November.
Cunningham, Alastair, Liz Dixon and Simon Hayes. 2001. Analysing yield spreads on emerging market sovereign bonds. Financial Stability Review, December.
Dam, Kenneth W. 1982. The Rules of the Game. Reform and Evolution in the International Monetary System. Chicago: University of Chicago Press.
Davis, E.P. 1992. Debt, Financial Fragility and Systemic Risk. Oxford: Clarendon Press.
Dell, Sidney. 1981. On Being Grandmotherly: The Evolution of Fund Conditionality. Essays in International Finance No. 144. New Jersey: Princeton.
Dell, Sidney. 1985. “The Fifth Credit Tranche”, World Development 13: 245-249.
Dell, Sidney. 1986. “The History of the IMF”, World Development 14: 1203-1212.
DeLong, John Bradford. 2000. “The Meltzer Report”. [www.j-bradford-delong.net]. May.
Eichengreen, Barry. 1999. Toward a New International Financial Architecture: A Practical Post- Asia Agenda. Washington, D.C.: Institute of International Economics.
Eichengreen, Barry. 2002. Financial crises and what to do about them. Oxford University Press: New York.
Eichengreen, Barry and Richard Portes. 1995. Crisis? What Crisis? Orderly Workouts for Sovereign Debtors., London: Centre for Economic Policy Research.
Eichengreen, Barry and Richard Portes. 2000. “A shortsighted vision for IMF reform”. Financial Times, 9 March.
Einhorn, Jessica. 2001. “The World Bank’s Mission Creep”. Foreign Affairs, September/October.
EMEPG. 2001. Building the International Financial Architecture. Emerging Markets Eminent Persons Group. Korea Institute for International Economics. Seul.
Fidler, Stephen. 2000. “Report urges slimming down of IMF and World Bank”. Financial Times, 8 March.
Fischer, Stanley. 1999. “On the need for an international lender of last resort”. Journal of Economic Perspectives 13: 23-37.
53 Gavin, Michael and Dani Rodrik 1995. “The World Bank in Historical Perspective”. American Economic Review Papers and Proceedings 85: 329-334.
Gilbert, Christopher, L. 2000. Comments on the Meltzer Report, FEWEC, Vrije Universiteit, Amsterdam.
Gilbert, Christopher, Andrew Powell, and David Vines. 1999. “Positioning the World Bank”. Economic Journal 109: 598-633.
Goldstein, Morris. 2000. “Strengthening the International Financial Architecture. Where Do We Stand?”. Working Paper 00-8. Institute of International Economics, Washington, D.C.
Group of 22. 1998. Report of the Working Group on International Financial Crises. Washington. D.C.
Haldane, Andrew G. 1999. “Private Sector Involvement in Financial Crisis: Analytics and Public Policy Approaches”. Financial Stability Review, November.
Haldane, Andrew G. and Jorg Scheibe. 2003. “IMF Lending and Creditor Moral Hazard”. Mimeo. Bank of England, July.
IDA (International Development Association). 2002. Addition to IDA Resources: Thirteen Replenishment. Report from the Executive Directors of IDA To the Board of Governors. Washington. D.C.: IDA.
IIF (Institute of International Finance). 2004. Capital Flows to Emerging Market Economies. January.
IMF. 1969. The International Monetary Fund 1945-1965, Vol. 1-3. Washington D.C.: IMF.
IMF. 2000. Executive Board discusses involving the private sector in the resolution of financial crisis. IMF Public Information Notice (PIN), no. 00/80. Washington, DC, 19 September.
IMF. 2003a. Proposals for a Sovereign Debt Restructuring Mechanism (SDRM). A Factsheet. January
IMF. 2003b. IMF Concludes Discussion on the Review of Contingent Credit Lines. Public Information Notice (PIN). No. 03/146. December 19.
IMF. 2004. IMF Financial Activities– Update March 26, 2004. www.imf.org/external/fin.htm.
Jungito, Roberto. 1994. “IMF-World Bank Policy Advice: The Coordination/Cross-Conditionality Question”. International Monetary and Financial Issues for the 1990s. Vol. IV. Geneva: UNCTAD.
54
Kaul, Inge, Isabelle Grunberg, and Marc A. Stern (eds.). 1999. Global Public Goods: International Cooperation in the 21st Century. New York: Oxford University Press.
Kapur, Davesh and Richard Webb. 1994. “The Evolution of Multilateral Development Banks”. International Monetary and Financial Issues for the 1990s. Vol. IV. Geneva: UNCTAD.
Kapur, Davesh, John Lewis and Richard Webb. 1997. The World Bank: Its First Half Century. Washington, D.C.: Brookings Institution.
Kapur, Davesh and Richard Webb. 2000. Governance-related Conditionalities of the International Financial Institutions. G24 Discussion Paper No. 6. Geneva: UNCTAD.
Kapur, Davesh. 2003. Do As I Say Not As I Do: A Critique of G-7 Proposals on Reforming the Multilateral Development Banks. G-24 Discussion Paper No. 20. Geneva: UNCTAD.
Kindleberger, Charles P. 1986. “International Public Goods without International Government”. American Economic Review, 76:1-13.
Kregel, Jan. (2004). External Financing for Development and International Financial Stability- A Balance Sheet Approach. Paper for G24 Meeting. Geneva.
Krueger, Anne 2001. International Financial Architecture for 2002: A New Approach to Sovereign Debt Restructuring, Address given at the National Economists’ Club Annual Members’ Dinner. Washington, DC.: American Enterprise Institute. November 26.
Krugman, Paul. 1996. “Are Currency Crises Self-Fulfilling?” In Ben S. Bernanke and Julio J. Rotemberg, eds. NBER Macroeconomics Annual: 345-506.
Krugman, Paul. 1998. “Currency Crises”. Unpublished paper. MIT, Mass.
Krugman, Paul. 1999. “Analytical Afterthoughts on the Asian Crisis”, unpublished paper. MIT, Mass.
Lerrick, Adam and Allan H. Meltzer. 2001. “The World Bank is Wrong to Oppose Grants”. Wall Street Journal, July 26.
Lerrick, Adam and Allan H. Meltzer. 2002. “Grants : A Better Way to Deliver Aid”. Quarterly International Economics Report. Carnegie Mellon Gailliot Center for Public Policy. January.
Mason, Edward S. and Robert E. Asher. 1973. The World Bank Since Bretton Woods. Washington, D.C.: The Brooking Institutions.
55 Meltzer, Allan H. 2001. “The Report of the International Financial Institution Advisory Commission: Comments on the Critics”. GSIA, Carnegie Mellon (Online).
Meltzer Commission. 2000. Report of the International Financial Institution Advisory Commission on Seven International Financial Institutions. Washington. D.C.: United States Congress.
Mikesell, Raymond F. 1994. The Bretton Woods Debates: A Memoir. Essays in International Finance, No. 192, March. Princeton; New Jersey.
Mina, Wasseem, and Jorge Martinez-Vazquez. 2002. “ IMF Lending, Maturity of International Debt and Moral Hazard”. Working Paper 03-01. Andrew Young School of Policy Studies.
Minsky, Hyman. 1977. “A Theory of Systemic Fragility”. In E. I Altman and A. W. Sametz, eds. Financial Crises. Wiley: New York.
Mishkin, Frederic S. 2004. The Economics of Money, Banking and Financial Markets. Pearson, Addison Wesley: Boston.
Obstfeld, Maurice. 1996. “Models of Currency Crises with Self-Fulfilling Features”. European Economic Review. 40: 1037-1047.
Ocampo, Jose Antonio. 2001. Recasting the International Financial Agenda. G24 Discussion Paper No. 13. Geneva: UNCTAD.
Oliver, Robert W. 1971. Early Plans for a World Bank. Princeton Studies in International Finance No. 29, Princeton: New Jersey.
Oliver, Robert W. 1975. International Economic Co-Operation and the World Bank. Macmillan: London.
Pesek, William Jr. (2003). “‘Bubble Reenters Asia Investors’ Lexicon”, Bloomberg.com, 22 December.
Phillips, Nicola and Richard Higgott. 1999. “Global Governance and the Public Domain: Collective Goods in a ‘Post-Washington Consensus Era”. CSGR W0rking Paper No. 47/99, University of Warwick.
Radelet, Steven. 1999. “Orderly Workouts for Cross-Border Private Debt”. International Monetary and Financial Issues for the 1990s. Vol. XI. Geneva: UNCTAD.
Ratha, Dilip. 2001. “Complementarity Between Multilateral Lending and Private Flows to Developing Countries: Some Empirical Results”. Mimeo, World Bank, Washington, D.C.
56 Rodrik, Dani. 1995. “Why is There Multilateral Lending?”. NBER Working Paper No. 5160.
Rwegasira, Delphin G. and Henock Kifle. 1994. “Regional Development Banks and the Objectives of the Bretton Woods Institutions”. International Monetary and Financial Issues for the 1990s. Vol. IV. Geneva: UNCTAD.
Sachs, Jeffrey D. 1995. “Do we need an international lender of last resort”, Frank D. Graham Lecture. Princeton University, Vol. 8, April 20.
Sachs, Jeffrey D. 1998. “External debt, structural adjustment and economic growth”. International Monetary and Financial Issues for the 1990s. Vol. IX. Geneva: UNCTAD.
Salazar, Vander Caceres. 2002. Taken for granted? Us Proposals to Reform the World Bank’s IDA Examined. London: Bretton Woods Project. March.
Sanford, Jonathan, E. 2002. World Bank: Funding IDA’s Assistance Program. Report for Congress, Order Code RL31418. Congressional Research Service Website.
Sanford, Jonathan, E. 2003. Multilateral Development Banks: issues for the 108th Congress. Issue Brief for Congress, Order Code IB96008. Washington D.C.: Congressional Research Service.
Stiglitz, Joseph E. 2002a. “Globalization and the Logic of Institutional Collective Action: Re- examining the Bretton Woods Institutions”. In Deepak Nayyar ed. Governing Globalization. New York: Oxford University Press.
Stiglitz, Joseph E. 2002b. Globalization and its Discontents. W. W. Norton & Company: New York.
Stiglitz, Joseph E. 2003. “Ethics, Market and Government failure, and Globalization”, paper presented to the Vatican Conference at the Ninth Plenary Session of the Pontifical Academy of Social Sciences, 2-6 May, Casina Pio IV.
Stiglitz, Joseph E. and Andrew Weiss. 1981. “Credit Rationing in Markets with Imperfect Information.” American Economic Review. 71: 393-410.
Summers, Lawrence H. 2000a. “The troubling aspects of IMF reform”. Financial Times, 23 March.
Summers, Lawrence H. 2000b. Testimony Before the Banking Committee of the House of Representatives. Treasury News , 23 March: Washington D.C.
57 Sy, Amadou N. R. 2001. Emerging Market Bond Spreads and Sovereign Credit Ratings: Reconciling Market View with Economic Fundamentals. IMF Working Paper WP/01/165.
Triffin, Robert. 1976. “Jamaica: “major revision” or fiasco”. In: E. M. Bernstein et al. Essays in International Finance, No. 115. New York: Princeton University, April.
UNCTAD. 1986. Trade and Development Report. Geneva: United Nations.
UNCTAD. 1987. “The Exchange Rate System”. In International Monetary and Financial Issues for the Developing Countries. Geneva: United Nations.
UNCTAD. 1998. Trade and Development Report. Geneva: United Nations.
UNCTAD. 1999. Trade and Development Report. Geneva: United Nations.
UNCTAD. 2000. Capital Flows and Growth in Africa. Geneva: United Nations.
UNCTAD. 2003. Trade and Development Report. Geneva: United Nations.
US Treasury. 2000. Response to the Report of the International Financial Institutions Advisory Commission. U.S. Department of Treasury. June. Washington, D.C..
Walters, Alan. 1994. Do We Need the IMF and the World Bank? London: Institute of Economic Affairs.
Williams, Ian. 2004. China-US: Double Bubbles in Danger of Colliding. Asia Times, 23 January.
Wolf, Martin. 2000. “Between Revolution and Reform – The Meltzer Commission’s Vision”. Financial Times, 8 March.
Woods, Ngaire. 1998. “Governance in international organizations: The case for reform in the Bretton Woods institutions”. In International Monetary and Financial Issues for the 1990s. Vol. IX. Geneva: UNCTAD.
World Bank. 2000. Global Development Finance. Washington, D.C.: World Bank
58
World Bank. 2001. World Bank Lending Instruments. Resources for Development Impact. Washington, D.C. : World Bank.
World Bank. 2002. Global Development Finance. Washington, D.C.: World Bank
Zedillo, Ernesto. 2001. Report of the High-level Panel on Financing for Development. United Nations: New York.
59
Table 1: World Bank Lending to Developing Countries (Billion Dollars)
1980 1990 1995 1996 1997 1998 1999 2000 2001 2002
Disbursements 6.1 17.8 18.7 19.5 20.5 20.0 19.5 18.6 18.4 17.6 IBRD 4.5 13.4 13.2 13.2 14.5 14.4 14.1 13.4 12.3 10.5 IDA 1.6 4.4 5.5 6.3 6.0 5.6 5.4 5.2 6.1 7.1
Net Flows 5.0 9.6 6.0 7.3 9.2 8.7 8.8 7.9 7.5 1.5 IBRD 3.4 5.5 1.0 1.6 3.9 3.9 4.3 3.6 2.5 -4.1 IDA 1.6 4.1 5.0 5.7 5.3 4.8 4.5 4.3 5.0 5.6
Net Transfers 3.1 2.4 -2.6 -0.9 2.0 1.5 0.9 -0.4 -0.6 -6.0 IBRD 1.6 -1.4 -7.0 -6.1 -2.8 -2.7 -3.0 -4.1 -5.0 -10.8 IDA 1.5 3.8 4.4 5.2 4.8 4.2 3.9 3.7 4.4 4.8
______
Source: World Bank Global Development Finance 2003. Net flows are equal to Disbursements minus Repayments; and Net Transfers are equal to Net Flows minus Interest Payments.